Evaluating the potential of urban development areas using artificial neural network Case Study: (Kermanshah City)

The rapid growth of urbanization and urban development, especially in developing countries, needs to be understood scientifically and efficiently by the complex patterns and processes of urban growth. In this study, the Multilayer Perceptron Neural Network (MLP) with the Levenberg-Marquardt Learning Algorithm for Kermanshah Urban Development Potential was used. Effective data in urban development were identified as inputs to the network. These layers are subdivided into three groups: socio-economic, land use and biophysical, including 16 layers: distance from urban areas, distance from city streets, distance from hospital, distance from the clinic, distance from the park and the green area, distance from main roads, distance between the commercial centers, distance from educational centers, distance from the fire, distance from the fault, distance from the gas station, agricultural use, the use of the mountain, the use of the plain, slope and elevation. The 500 points were provided as network teaching points and the number of layers was 12 layers. Finally, according to the results, with the removal of facilities and urban areas, the potential has declined sharply, and most of the regions have urban development potential at the closest distance between these facilities and urban areas. The most potential areas of urban development are located in the southwest of Kermanshah and around the main roads of Kermanshah-Islamabad and Kermanshah-Kengavar. The northern areas of the city have a low development potential due to their height and slope.